Physical laws combined with empirical common sense form the basis for most water engineering applications. For example, in the specification of rainfall-runoff models, estimation of design floods, study of sediment movement in rivers, evaluation of solute transport in groundwater, planning and management of water resources systems, and assessment of impacts of climate change on water resources, little is accomplished without a sound basis for linking empirical knowledge with laws of science. This session focuses on the role of empirical or data-based techniques in water engineering. These include, among others, approaches for quantifying parameters or model uncertainty (in hydrologic and non-hydrologic models), stochastic approaches (with applications to generation of hydrologic time series, downscaling of global climate model outputs to hydrologic variables at the catchment scale, continuous simulation approaches for design flood estimation or planning studies requiring time series inputs), nonlinear deterministic approaches (for data reconstruction, pattern recognition, dimension reduction, and model simplification in hydrology and water resources), complex networks (for identifying structures, connections, and spatio-temporal dynamics of systems), and other data-based or data-driven modeling techniques (with applications to short-, medium- or long-term forecasting of hydrologic variables). Submissions are welcome from all streams of water engineering by researchers and practitioners with an interest in data and data-based techniques. Studies addressing integration of different techniques are particularly encouraged.